JP2006127201A - Storage system and conduction confirmation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a conduction confirmation technique for facilitating conduction confirmation of a data transfer path in an IP storage system. <P>SOLUTION: A target computer 20 of a storage system 10 transmits an ICMP echo request to an initiator computer 70 via a prescribed IP port 110 provided by the storage system 10, to determine whether the conduction between the initiator computer 70 and the target computer 20 via the prescribed IP port 110 is normal based on a reception status of the ICMP echo response from the initiator computer 70. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a continuity confirmation technique for confirming continuity on an IP network in a storage system that transfers storage data via an Internet Protocol (IP) network.

  In recent years, IP storage systems that transfer storage data via an IP network have been put to practical use as storage systems. Technologies for realizing IP storage systems include iSCSI (Internet Small Computer System Interface), which maps SCSI commands to IP, and multiple independent FC-LANs (Fibre Channel-Local Area Networks) via IP tunnels. Technologies such as FCIP (Fibre Cannel over IP), iFCP (Internet Fiber Channel Protocol) that performs routing by mapping an FC (Fibre Cannel) address and an IP address are known. For example, in IP storage based on the iSCSI protocol, storage data is transferred via the IP network by executing an iSCSI session on a TCP (Transmission Control Protocol) connection via the IP network.

  In general, an IP storage system includes a plurality of connection ports that can be connected to an IP network, and a target computer that provides a storage volume (storage area) to the initiator computer connected to the IP network through a predetermined connection port. A predetermined connection port as a data transfer path between the initiator computer and the target computer is preset on the target computer side, and the initiator computer passes through the predetermined connection port preset as a data transfer path. A storage volume is provided from the target computer by establishing a communication session for. Patent Document 1 below discloses an IP storage system that transfers storage data via an IP network.

JP 2003-345631 A

  However, conventionally, after a predetermined connection port has been set as a data transfer path on the IP storage system side, the continuity of the data transfer path has been confirmed by attempting access from the initiator computer side according to the setting. When there is some kind of access failure in the transfer path, there is a problem that it is compelling to repeat the setting of the data transfer path on the IP storage system side and the conduction check on the initiator computer side. Examples of data transfer path access failures include failure of relay devices on the IP network, path setting errors on the relay device side of the IP network, access restrictions due to firewalls, excessive communication load, connection port There may be a failure.

  In view of the above-described problems, an object of the present invention is to provide a continuity confirmation technique for facilitating confirmation of continuity of a data transfer path in an IP storage system.

  In order to solve the above-described problem, the continuity confirmation method of the present invention includes a plurality of connection ports connectable to a network based on the Internet protocol, and at least one of the plurality of connection ports for an initiator computer connected to the network. A continuity confirmation method for confirming continuity between the initiator computer and the target computer in a storage system including a target computer that provides a predetermined storage volume through an Internet control message protocol by the target computer An echo request step for transmitting an echo request based on the request to the initiator computer through a predetermined connection port of the plurality of connection ports; and Based on the reception state at the target computer of the echo response returned from the initiator computer as an answer, the continuity between the initiator computer and the target computer through the predetermined connection port is normal by the target computer And a continuity determining step for determining whether or not.

  In order to solve the above-described problems, a storage system according to the present invention includes a plurality of connection ports connectable to a network based on an Internet protocol, and at least one of the plurality of connection ports with respect to an initiator computer connected to the network. A storage system including a target computer that provides a predetermined storage volume, wherein the target computer sends an echo request based on an Internet control message protocol through the predetermined connection port of the plurality of connection ports. Echo request means for transmitting to the computer, and reception of the echo response returned from the initiator computer as a response to the transmitted echo request at the target computer. Based on the state, characterized in that the conduction between the predetermined said initiator computer and the target computer through the connection port and a conduction determining means for determining whether or not normal.

  According to the continuity confirmation method and the storage system of the present invention, whether or not the continuity of the data transfer path is normal by an echo request based on the Internet Control Message Protocol (ICMP) from the target computer on the storage system side. Therefore, it is not necessary to attempt access from the initiator computer side for confirmation of continuity, and it is possible to facilitate confirmation of continuity of the data transfer path.

  The continuity confirmation method and the storage system of the present invention described above can also take the following aspects. In the conduction check method of the present invention, the predetermined connection port is a connection port newly set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports, A path for registering a setting for using the connection port as the data transfer path by the target computer when it is determined that conduction between the initiator computer and the target computer through the predetermined connection port is normal. A setting registration step may be provided. In the storage system of the present invention, the predetermined connection port is a connection port newly set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports, and the target Further, when the computer determines that continuity between the initiator computer and the target computer through the predetermined connection port is normal, a path setting for registering a setting using the connection port as the data transfer path is registered. Registration means may be provided. According to these aspects of the invention, it is possible to facilitate confirmation of continuity when a data transfer path is newly set.

  Further, in the continuity check method of the present invention, the predetermined connection port is a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports. When a communication failure occurs in a connection port that is a different connection port and that is set as the data transfer path, the continuity of the connection port set as the data transfer path by the target computer is normal. A path setting changing step for changing the setting to at least one of the connection ports determined to be may be provided. Further, when changing the setting of the connection port in the path setting changing step, a port address indicating the location of the connection port on the network, which has been assigned to the connection port of the setting change source, is used as the connection port of the setting change destination. An address transfer process for transferring may be provided. In the storage system of the present invention, the predetermined connection port is a connection port different from a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports. The target computer further has a normal connection to the connection port set as the data transfer path when a communication failure occurs in the connection port set as the data transfer path. Path setting change means for changing the setting to at least one of the determined connection ports may be provided. The target computer further changes the port address indicating the location of the connection port on the network that was assigned to the connection port of the setting change source when the connection port setting is changed by the path setting changing unit. Address transfer means for transferring to the previous connection port may be provided. According to the inventions of these aspects, even if a communication failure occurs in the connection port set as the data transfer path, the data transfer path can be quickly recovered. Furthermore, by transferring the port address, the data transfer path can be recovered without changing the communication setting on the initiator computer side.

  Further, in the continuity check method of the present invention, the predetermined connection port is a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports. When the communication load of the connection port that is a different connection port and is set as the data transfer path exceeds a predetermined value, the connection port that is set as the data transfer path by the target computer is A path setting change process for changing the setting to at least one of the connection ports determined to have normal conduction may be provided. Further, when changing the setting of the connection port in the path setting changing step, a port address indicating the location of the connection port on the network, which has been assigned to the connection port of the setting change source, is used as the connection port of the setting change destination. An address transfer process for transferring may be provided. In the storage system of the present invention, the predetermined connection port is a connection port different from a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports. The target computer further connects the connection port set as the data transfer path to the normal connection when the communication load of the connection port set as the data transfer path exceeds a predetermined value. It is also possible to provide path setting changing means for changing the setting to at least one of the connection ports determined to be. Further, the target computer further includes a port address indicating a location on the network of the connection port assigned to the connection port of the setting change source when the connection port setting is changed by the path setting changing unit. Address transfer means for transferring to the connection port of the setting change destination may be provided. According to the inventions of these aspects, even when the communication load of the connection port set as the data transfer path is excessive, it is possible to quickly shift to a data transfer path having a good communication state. . Furthermore, by transferring the port address, the data transfer path can be recovered without changing the communication setting on the initiator computer side.

  Further, in the conduction confirmation method of the present invention, the predetermined connection port is a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports, The target computer is capable of exchanging data with a management terminal that exchanges predetermined information with an administrator of the storage system, and further, from the administrator via the management terminal by the target computer. A determination result notifying step of notifying the administrator via the management terminal of the determination result as to whether or not the continuity of the predetermined connection port is normal based on the input of the instruction Also good. In the storage system of the present invention, the predetermined connection port is a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports, and the target computer Is capable of exchanging data with a management terminal that exchanges predetermined information with the administrator of the storage system, and further, based on an instruction input from the administrator via the management terminal A determination result notifying unit for notifying the administrator of the determination result as to whether or not the continuity of the predetermined connection port is normal may be provided. According to the inventions of these aspects, the administrator of the storage system can easily confirm the continuity with respect to the existing data transfer path, and the maintenance management of the storage system can be facilitated. The management terminal may be connected to the storage system via a network, or may be incorporated in the storage system.

  Further, in the continuity confirmation method of the present invention, the continuity determination step may be a step of determining that the continuity is normal when the echo response reaches the target computer. In the storage system according to the present invention, the continuity determining means may be means for determining that the continuity is normal when the echo response reaches the target computer. According to the inventions of these aspects, the processing speed of the continuity determination can be improved without complicated determination processing.

  Further, in the continuity confirmation method of the present invention, the continuity determination step is performed when the arrival rate at which the echo response reaches the target computer exceeds a predetermined value in response to a plurality of the echo requests. It is good also as a process of judging that it is. In the storage system according to the present invention, the continuity determination unit may determine that the continuity is normal when an arrival rate at which the echo response reaches the target computer exceeds a predetermined value in response to a plurality of the echo requests. It may be a step of judging. According to the inventions of these aspects, it is possible to prevent a data transfer path having an excessive communication load from being determined to be normal.

  In the continuity check method of the present invention, the target computer can exchange data with a user output interface that outputs predetermined information to an administrator of the storage system, and further, the predetermined connection An abnormality notification step of notifying the administrator of the reason for the abnormality via the user output interface by the target computer when it is determined that conduction between the initiator computer and the target computer through the port is abnormal It is good also as having. In the storage system of the present invention, the target computer can exchange data with a user output interface that outputs predetermined information to an administrator of the storage system, and further through the predetermined connection port. In addition, when it is determined that the continuity between the initiator computer and the target computer is abnormal, an abnormality notification unit is provided for notifying the administrator of the reason for the abnormality via the user output interface. Also good. According to these aspects of the invention, the administrator of the storage system can take measures for restoring the data transfer path based on the notified reason for the abnormality. The user output interface may be connected to the storage system via a network or may be incorporated in the storage system.

  Further, in the continuity confirmation method of the present invention, the target computer can exchange data with a user input interface that receives input of predetermined information from an administrator of the storage system, and further, by the target computer, An address acquisition step of acquiring an initiator address indicating the location of the initiator on the network from the administrator via the user input interface may be provided. In the storage system of the present invention, the target computer can exchange data with a user input interface that receives input of predetermined information from an administrator of the storage system, and further via the user input interface Address acquisition means for acquiring an initiator address indicating the location of the initiator on the network from the administrator may be provided. According to these aspects of the invention, it is possible to facilitate the confirmation of continuity when setting the data transfer path by inputting the initiator address by the administrator of the storage system. The user input interface may be connected to the storage system via a network, or may be built into the storage system.

  Further, in the continuity confirmation method of the present invention, the target computer associates an initiator name that can identify the initiator computer with a storage name server that holds an initiator address indicating the location of the initiator on the network, and It is possible to exchange data with a user input interface that receives input of predetermined information from an administrator of the storage system, and further, the target computer acquires the initiator name from the administrator via the user input interface, An address acquisition step of acquiring the initiator address held by the storage name server in association with the acquired initiator name from the storage name server may be provided. The storage system according to the present invention, wherein the target computer is associated with an initiator name that can identify the initiator computer, and stores an initiator address indicating the location of the initiator on the network, and the storage system Data can be exchanged with a user input interface that accepts input of predetermined information from an administrator, and the initiator name is acquired from the administrator via the user input interface and associated with the acquired initiator name. It is also possible to provide address acquisition means for acquiring the initiator address held by the storage name server from the storage name server. According to the inventions of these aspects, it is possible to facilitate the confirmation of continuity when setting the data transfer path by inputting the initiator name by the administrator of the storage system. The storage name server may be connected to the storage system via a network or may be incorporated in the storage system. The various data held by the storage name server is data registered in advance by the administrator of the storage system. In addition, the user input interface may be connected to the storage system via a network or may be incorporated in the storage system.

  In the continuity check method of the present invention, the target computer includes a storage domain that defines a combination of the initiator computer and the target computer, and the initiator computer defined in the storage domain on the network. The initiator address of the initiator computer that is capable of exchanging data with a storage name server having an initiator address indicating the location and that is defined in the storage domain as a combination with the target computer by the target computer May be provided with an address acquisition step of acquiring from the storage name server. In the storage system of the present invention, the target computer indicates a storage domain in which a combination of the initiator computer and the target computer is defined, and the location of the initiator computer defined in the storage domain on the network Data can be exchanged with a storage name server having an initiator address, and the initiator address of the initiator computer defined in the storage domain as a combination with the target computer can be transferred from the storage name server. Address acquisition means for acquiring may be provided. According to the inventions of these aspects, it is possible to facilitate continuity confirmation when setting the data transfer path according to the storage name server. The storage name server may be connected to the storage system via a network or may be incorporated in the storage system. The various data held by the storage name server is data registered in advance by the administrator of the storage system.

  Further, at least one of the initiator computer and the target computer may be plural. According to the invention of this aspect, it is possible to facilitate the confirmation of continuity of the data transfer path having a huge number of setting patterns. The communication between the initiator computer and the target computer may be communication based on the iSCSI protocol.

  In order to further clarify the configuration and operation of the present invention described above, the conduction confirmation technique to which the present invention is applied will be described in the following order.

Table of Contents A. Configuration of electronic computer system CS A (1). Physical configuration of electronic computer system CS A (2). B. Logical configuration of electronic computer system CS Operation of electronic computer system CS B (1). New path conduction confirmation process B (2). Processing for confirming continuity of existing path B (3). B. Path restoration processing in case of failure / high load Other embodiments

A. Configuration of the electronic computer system CS:
A (1). Physical configuration of the electronic computer system CS:
FIG. 1 is an explanatory diagram showing a physical configuration of the electronic computer system CS. An electronic computer system CS that is one embodiment of the present invention includes a storage system 10 that constitutes a storage volume (hereinafter referred to as a volume) that is a storage area capable of storing data, and an initiator that uses the volume of the storage system 10. The computer 70 and an iSNS (internet Storage Name Server) server 50 that manages the relationship between the storage system 10 and the initiator computer 70 are provided. Each of the storage system 10, the initiator computer 70, and the iSNS server 50 is connected to each other by a plurality of IP switches 610 that transfer IP packets, and can exchange data with each other via an IP network. Storage Area Network) 60 is configured. Each of the storage system 10, the initiator computer 70, the iSNS server 50, and the IP switch 610 is a device that supports various network protocols such as an IP protocol, a TCP protocol, and an ICMP protocol. The storage system 10 and the initiator computer 70 are configured to be operable as devices compatible with the iSCSI protocol, which is a protocol for realizing storage data transfer via the IP-SAN 60.

  FIG. 2 is a block diagram mainly showing the internal configuration of the storage system 10. The storage system 10 includes a storage unit 30 that logically configures a volume by a plurality of IP ports 110 that can be connected to the IP-SAN 60, a plurality of hard disk drives (hereinafter referred to as HDDs) 310, and a storage unit 30. A target computer 20 that provides the initiator computer 70 with at least one of the plurality of IP ports 110 and a LAN port 130 that connects the target computer 20 to a LAN (Local Area Network). In this embodiment, the storage system 10 includes 16 IP ports 110, and each of the 16 IP ports 110 is assigned a port address that is an IP address indicating a location on each IP-SAN. Yes.

  The target computer 20 of the storage system 10 stores in advance a central processing unit (hereinafter referred to as a CPU) 210 that performs arithmetic processing for controlling each part of the target computer 20 and a program that defines the arithmetic processing of the CPU 210. A read only memory (hereinafter referred to as ROM) 212, a random access memory (hereinafter referred to as RAM) 214 that temporarily stores data handled by the CPU 210, a plurality of IP ports 110 and a plurality of A data controller 220 having a circuit for controlling exchange of storage data with the HDD 310 and a cache memory 230 for temporarily storing data handled by the data controller 220 are provided.

  The ROM 212 of the target computer 20 stores an operating system (hereinafter referred to as OS) and various application software (hereinafter referred to as applications) as programs executed by the CPU 210. Examples of various applications stored in the ROM 212 include ping (Packet Internet Groper), which is a program for diagnosing an IP network based on the ICMP protocol, and a plurality of HDDs 310 in the storage unit 30 using RAID (Redundant Arrays of Independent (Inexpensive). ) Disks, Raid) There are programs to control using technology. Note that these programs may be stored in an external storage device such as an HDD, and loaded into the RAM 214 and executed at startup.

  A management terminal 15 having various user interfaces such as a display 16 and a keyboard 17 is connected to the LAN port 130 of the storage system 10, and the storage control unit 20 communicates with the administrator of the storage system 10 via the management terminal 15. Exchange of predetermined information is possible.

  The initiator computer 70 is a computer equipped with hardware such as a CPU, ROM, RAM, HDD, and various interface circuits. The initiator computer 70 is installed with various applications in addition to the OS. In the initiator computer 70, an application for executing data processing, an application for using the storage system 10, and the like are installed.

  The iSNS server 50 is a computer that includes hardware such as a CPU, ROM, RAM, HDD, and various interface circuits. The iSNS server 50 has various applications installed in addition to the OS. The iSNS server 50 includes a storage domain in which the relationship between the storage system 10 and the initiator computer 70 is defined as information for the storage system 10 and the initiator computer 70 to specify a communication partner, and the storage system 10 and the initiator computer 70. Information such as an IP address indicating the location on the IP-SAN 60 is registered. A management terminal 15 having various user interfaces such as a display 56 and a keyboard 57 is connected to the iSNS server 50, and the iSNS server 50 exchanges predetermined information with an administrator of the iSNS server 50 via the management terminal 15. Is possible.

A (2). Logical configuration of the electronic computer system CS:
FIG. 3 is an explanatory diagram showing a logical configuration of the electronic computer system CS. In the computer system CS, the initiator computer 70 logically configures an iSCSI initiator, and the target computer 20 of the storage system 10 logically configures an iSCSI target. In the example shown in FIG. 3, three initiators IT1 to IT3 are logically configured as iSCSI initiators on the three initiator computers 70, and two targets TA1 and TA2 are set as iSCSI targets on the target computer 20 of the storage system 10. Logically structured.

  FIG. 4 is an explanatory diagram illustrating an example of the target setting table. The target setting table 4000 shown in FIG. 4 is information for setting the volume allocation of the storage unit 30 to the iSCSI target of the target computer 20 and is stored in the RAM 214 built in the target computer 20. The target setting table 4000 includes a target name 4010 for identifying an iSCSI target configured in the storage system 10, a volume name 4020 for identifying a volume configured in the storage system 10, and a storage capacity 4030 assigned to each volume. Each information is included. In the example shown in FIG. 3, based on the target setting table 4000 shown in FIG. 4, among the three volumes BO1 to 3 each having a storage capacity of 10 gigabytes, the volumes BO1 and BO2 are assigned to the target TA1, and the volume BO3 Is assigned to the target TA2.

  FIG. 5 is an explanatory diagram showing an example of the path setting table. The path setting table 5000 shown in FIG. 5 is information for setting an iSCSI session between the iSCSI initiator of the initiator computer 70 and the iSCSI target of the target computer 20, and is stored in the RAM 214 built in the target computer 20. Yes. The path setting table 5000 includes a target name 5010, an initiator name 5020 for identifying an iSCSI initiator that permits an iSCSI session with the iSCSI target, an initiator address 5030 that is an IP address of the iSCSI initiator, and between the iSCSI initiator and the iSCSI target. Information of IP port name 5040 for identifying IP port 110 as a data transfer path (path) used for establishing TCP connection, TCP port number 5050 used for each TCP connection, authentication information 5060 used for authentication of iSCSI initiator including. In the present embodiment, each of the 16 IP ports 110 is assigned the IP port name of the ports PO1 to PO16. In the example shown in FIG. 3, ports PO1 and PO2 are assigned as data transfer paths between the target TA1 and the initiators IT1 and IT2 based on the path setting table 5000 shown in FIG. A port PO2 is assigned as a data transfer path between them.

B. Operation of the electronic computer system CS:
B (1). New path continuity confirmation process:
FIG. 6 is a flowchart showing the new path setting process. The new path setting process is a process for registering a new path, which is a data transfer path to be newly set, in the path setting table 5000 shown in FIG. 5, and in this embodiment, software of the CPU 210 in the target computer 20 This process is realized by an operation based on the above. In this embodiment, the target computer 20 starts a new path setting process based on an instruction from the administrator via the management terminal 15.

  When the target computer 20 starts the new path setting process shown in FIG. 6, the target computer 20 specifies the iSCSI target that is the target for setting the new path as the target target (step S110). In the present embodiment, the target computer 20 specifies the iSCSI target designated by the administrator via the management terminal 15 as the target target.

  After specifying the target target (step S110), the target computer 20 specifies the iSCSI initiator that is the target of setting a new path with the target target as the target initiator, and indicates the location of the specified target initiator on the IP-SAN 60. Initiator information acquisition processing is executed to acquire an initiator address that is an IP address (step S120). Details of the initiator information acquisition process will be described later.

  After executing the initiator information acquisition process (step S120), the target computer 20 specifies the IP port 110 set as a new path between the target target and the target initiator as the target port (step S130). In the present embodiment, the target computer 20 identifies the IP port 110 designated by the administrator via the management terminal 15 as the target port.

  After identifying the target port (step S130), the target computer 20 sets a new path connecting the target initiator and the target target through the target port as a continuity confirmation path for performing continuity confirmation (step S140). Thereafter, the target computer 20 executes a continuity confirmation process by an ICMP echo request for the set continuity confirmation path (step S150).

  FIG. 7 is a flowchart showing the continuity confirmation process. When starting the continuity confirmation process, the target computer 20 transmits an ICMP echo request for the initiator address of the target initiator through the target port by executing a ping command installed as an application (step S210). Thereafter, it is determined whether or not an ICMP echo response to the transmitted ICMP echo request has been received (step S220). If an ICMP echo response has been received, it is determined that the continuity of the continuity confirmation path is normal (step S230). ). On the other hand, when a predetermined time elapses after the transmission of the ICMP echo request, the reception waiting time of the ICMP echo response times out, and when the ICMP echo response cannot be received, the conduction of the conduction confirmation path is abnormal. Judgment is made (step S240). Thereafter, when continuity confirmation for all continuity confirmation paths is completed (step S250), the continuity confirmation process is terminated. It should be noted that when the ICMP echo response reception waiting time has timed out, the transmission of the ICMP echo request may be retried to determine whether or not the ICMP echo response has been received. When the ICMP echo request does not reach the destination, the target computer 20 receives the ICMP message indicating the cause of the error such as non-delivery of the destination or excess of time transmitted from the IP switch 610 in the path route. The reason for the continuity abnormality can be determined.

  FIG. 8 is an explanatory diagram showing an example of the state of the continuity confirmation process. The example shown in FIG. 8 shows the state of the continuity confirmation process when the path setting related to the target TA1 in the path setting table 5000 shown in FIG. 5 is set and registered as a new path. In the example shown in FIG. 8, the path between the target TA1 and the initiator IT1 through the port PO1, the path between the target TA1 and the initiator IT1 through the port PO2, and the path between the target TA1 and the initiator IT2 through the port PO1. An ICMP echo request is transmitted for each of the path and the path between the target TA1 and the initiator IT2 through the port PO2, and the reception state of the ICMP echo response is determined.

  Returning to the description of the new path setting process in FIG. 6, the target computer 20 ends the continuity confirmation process shown in FIG. 7 (step S150), and when the continuity of all continuity confirmation paths is normal (step S150). In step S155), a new path is registered in the path setting table 5000 shown in FIG. 5 (step S160). If the path is not normal, the new path is not registered. After that, the continuity confirmation result of the new path is notified to the administrator via the management terminal 15 (step S170).

  FIG. 9 is an explanatory diagram illustrating an example of notification of a conduction confirmation result of a new path. FIG. 9 shows a result display window WD1 displayed on the display 16 of the management terminal 15 as a notification of the continuity confirmation result of the new path. In the result display window WD1, information on whether or not the conduction state is normal is displayed for each path on which the conduction confirmation process is performed. If the conduction state is abnormal, the determination is made from the ICMP message indicating the cause of the error. The reason for the error is displayed. The target computer 20 notifies the conduction confirmation result of the new path (step S170), and then ends the new path setting process.

  FIG. 10 is a flowchart showing initiator information acquisition processing. The initiator information acquisition process is a process (step S120 in FIG. 6) executed in the new path setting process shown in FIG. 6, and is a process for specifying the target initiator and acquiring the initiator address. . When an initiator name that can identify the iSCSI initiator and its initiator address are input from the administrator via the management terminal 15, the target computer 20 identifies the iSCSI initiator designated by the administrator as the target initiator. The initiator address input from the administrator is acquired (steps S320 and S340).

  On the other hand, when the initiator name is not input or when the initiator address is not input, the target computer 20 connects to the iSNS server 50 via the IP-SAN 60 and acquires necessary information from the iSNS server 50. FIG. 11 is an explanatory diagram of an example of the storage domain setting table. FIG. 12 is an explanatory diagram showing an example of the initiator address table. The storage domain setting table 11000 shown in FIG. 11 is information in which a storage domain related to the iSCSI target of the storage system 10 is set. The storage domain name 11010 for identifying the storage domain and the nodes included in each storage domain are identified. For each node name 11020, and an attribute 11030 indicating whether each node is an initiator or a target. The initiator address table 12000 shown in FIG. 12 is information in which the initiator address of the iSCSI initiator registered in the storage domain setting table 11000 is recorded. Each information of the initiator name 12010 of each iSCSI initiator and the initiator address 12020 of each iSCSI initiator. including. In this embodiment, the storage domain setting table 11000 and the initiator address table 12000 are registered in the iSNS server 50 from the administrator via the management terminal 55.

  When the initiator name is not input from the administrator (step S310), the target computer 20 is defined in the storage domain setting table 11000 as a combination with the target target by presenting the target target to the iSNS server 50. The iSCSI initiator is specified as the target initiator, and the initiator address of the target initiator stored in the initiator address table 12000 is acquired from the iSNS server 50 (step S360).

  When the target computer 20 receives the initiator name from the administrator but does not receive the initiator address (step S330), the target computer 20 is stored in the initiator address table 12000 by presenting the target initiator to the iSNS server 50. The initiator address of the target initiator is acquired from the iSNS server 50 (step S350).

  After the target computer 20 identifies the target initiator and obtains the initiator address (steps S320, S340, S350, S360), the target computer 20 associates the initiator address with the initiator name in the RAM 214 or other storage (not shown). (Step S350), and the initiator information acquisition process is terminated.

B (2). Processing for confirming continuity of existing path:
FIG. 13 is a flowchart showing the existing path confirmation process. The existing path setting process is a process for performing continuity confirmation on an existing path which is a set data transfer path. In this embodiment, the process is realized by an operation based on the software of the CPU 210 in the target computer 20. It is.

  When the target computer 20 receives an instruction to start processing from the administrator via the management terminal 15, the target computer 20 starts existing path confirmation processing (step S310), and the existing path set in the path setting table 5000 shown in FIG. It is set as a conduction confirmation path for conducting conduction confirmation (step S320). Thereafter, the target computer 20 executes the continuity confirmation process shown in FIG. 7 (step S330), and then notifies the administrator of the continuity confirmation result of the existing path via the management terminal 15 (step S340). FIG. 14 is an explanatory diagram showing an example of notification of the result of confirming the continuity of the existing path. FIG. 14 shows a result display window WD2 displayed on the display 16 of the management terminal 15 as a notification of the result of confirming the continuity of the existing path. In the result display window WD2, information on whether or not the continuity state is normal is displayed for each path on which the continuity confirmation processing is performed. The reason for the error is displayed. The target computer 20 notifies the continuity confirmation result of the existing path (step S340), and then ends the existing path setting process.

B (3). Path recovery processing in the event of failure or high load:
FIG. 15 is a flowchart showing the path recovery process. The path recovery process is a process for recovering a path that has become difficult to use due to a failure or a high load. In this embodiment, the path recovery process is a process realized by an operation based on the software of the CPU 210 in the target computer 20. In the present embodiment, the target computer 20 repeatedly executes the path restoration process at a predetermined timing.

  When the target computer 20 starts the path recovery process, the target computer 20 has a faulty port in which a fault has occurred or a high load among the IP ports 110 set in the path setting table 5000 shown in FIG. A high load port is searched (step S420). FIG. 16 is an explanatory diagram of an example of the IP port communication statistics table. The IP port communication statistics table 16000 is information in which the target computer 20 records the continuity state and the load factor for each of all the IP ports 110 included in the storage system 10 and is stored in the RAM 214 built in the target computer 20. . The IP port communication statistics table 16000 includes information on the IP port name 16010 of each IP port 110, the conduction state 16020 of each IP port 110, and the load factor 16030 of each IP port 110. The target computer 20 searches for a fault / high load port by referring to the IP port communication statistics table 16000 stored in the RAM 214. In this embodiment, if the conduction state is blocked, it is determined as a faulty port, and if the load factor exceeds 80%, it is determined as a high load port.

  When there is a failure / high load port (step S425), the target computer 20 refers to the IP port communication statistics table 16000 shown in FIG. Then, an unused port that is not set in the path setting table 5000 shown in FIG. 5 is searched (step S430).

  If there is an unused port (step S435), the target computer 20 replaces the faulty / high load port of the existing path with an unused port based on the existing path using the fault / high load port. Then, it is set as a conduction path for conducting conduction confirmation (step S440). After that, the target computer 20 executes the continuity confirmation process shown in FIG. 7 (step S450). If the continuity of the continuity confirmation path is normal (step S455), the target computer 20 uses the fault / high load port. By registering an alternative path in place of the path in the path setting table 5000 shown in FIG. 5, the setting of the existing path is changed to the alternative path (step S460).

  FIG. 17 is an explanatory diagram of an example of the port address table. The port address table 17000 is information in which the port addresses of the 16 IP ports 110 included in the storage system 10 are recorded, and is stored in the RAM 214 built in the target computer 20. The port address table 17000 includes information on the IP port name 17010 of each IP port 110 and the port address 17020 assigned to each IP port 110. After changing the setting of the existing path to the alternative path (step S460), the target computer 20 transfers the port address assigned to the setting change source IP port 110 to the setting change destination IP port 110, as shown in FIG. The port address table 17000 shown in FIG. 4 is rewritten (step S470).

  The target computer 20 performs processing for all fault / high load ports (step S475), and then notifies the administrator of the path recovery processing result via the management terminal 15 (step S480). FIG. 18 is an explanatory diagram showing an example of notification of the path recovery processing result. FIG. 18 shows a result display window WD3 displayed on the display 16 of the management terminal 15 as a notification of the path restoration processing result. The result display window WD3 displays that the communication status is good for the existing path that does not use the fault / high load port, and the setting change is displayed for the path that has been changed to the alternative path. For a path whose setting could not be changed to an alternative path, a message to that effect is displayed. After notifying the path recovery processing result (step S480), the target computer 20 ends the path recovery process.

  According to the storage system 10 of the present invention described above, the storage system 10 can determine whether or not the continuity of the data transfer path is normal based on the ICMP echo request from the target computer 20 on the storage system 10 side. Therefore, it is not necessary to attempt access from the initiator computer 70 side for confirming the continuity, and the continuity confirmation of the data transfer path can be facilitated.

  In addition, the new path setting process shown in FIG. 6 can facilitate continuity confirmation when a new data transfer path is set. Further, the existing path confirmation process shown in FIG. 13 allows the administrator of the storage system 10 to easily confirm the continuity of the existing data transfer path, thereby facilitating the maintenance management of the storage system 10. Can do. Further, even when the IP port 110 set as the data transfer path becomes a failure / high load port by the path recovery process shown in FIG. 15, the data transfer path has a good communication state. You can move quickly. Further, by transferring the port address (step S470 in FIG. 15), the data transfer path can be recovered without changing the communication setting on the initiator computer 70 side.

C. Other embodiments:
As mentioned above, although embodiment of this invention was described, this invention is not limited to such embodiment at all, Of course, it can implement with various forms within the range which does not deviate from the meaning of this invention. is there. For example, in the continuity confirmation process shown in FIG. 7, it is determined whether or not continuity is normal depending on whether or not an ICMP echo response is received. However, whether or not continuity is normal depending on the arrival rate of the ICPM echo response. It is good also to judge. FIG. 19 is a flowchart illustrating a continuity confirmation process according to another embodiment. In the continuity confirmation process shown in FIG. 19, the target computer 20 transmits a plurality of ICMP echo requests (step S510), and then determines whether or not the arrival rate at which the ICMP echo response reaches the target computer 20 exceeds a predetermined value. (Step S520). When the arrival rate of the ICMP echo response exceeds a predetermined value, the target computer 20 determines that the conduction of the conduction confirmation path is normal (step S530), and the arrival rate of the ICMP echo response does not exceed the predetermined value. Is determined to be abnormal in the conduction confirmation path (step S540). In this embodiment, the predetermined value of the arrival rate of the ICMP echo response that is a criterion for determining continuity is 90%, but this value can be appropriately changed according to the embodiment. It is possible to prevent a path having an excessive communication load from being determined to be normal by the continuity confirmation process of the other embodiment illustrated in FIG.

  In addition, when setting communication with the iSNS server 50 via the IP-SAN 60 in the storage system 10, the target computer 20 performs the continuity confirmation process shown in FIG. 7 or 19 with the iSNS server 50. It may be executed for the data transfer path. At this time, the target computer 20 may execute continuity confirmation processing for paths to the iSNS server 50 through all IP ports 110. FIG. 22 is a flowchart showing server connection setting processing. The server connection setting process shown in FIG. 22 is a process executed when registering the iSNS server address, which is the IP address of the iSNS server 50, in the target computer 20. After receiving the iSNS server address input from the user via the management terminal 15 (step S710), the target computer 20 sets a path to the iSNS server 50 through all the IP ports 110 as a continuity confirmation path (step S710). S720), the conduction confirmation process shown in FIG. 7 or FIG. 19 is executed (step S730). Thereafter, the target computer 20 determines whether or not the continuity result of each IP port 110 in the continuity confirmation process is within an allowable range for registering the iSNS server address (step S740). In this embodiment, the determination criterion for determining that it is within the allowable range is to permit registration if half of all the IP ports 110 are normal, but the determination criterion can be set as appropriate. Registration may be permitted when conduction of all IP ports 110 is normal. When the target computer 20 determines that the continuity result is within an allowable range (step S740), the iSNS server address and the continuity result are registered in a predetermined table stored in the RAM 214 (steps S750 and S760), and the server connection setting result Notification is performed (step S770). FIG. 23 is an explanatory diagram of an example of notification of a server connection setting result. FIG. 23 shows a result display window WD4 displayed on the display 16 of the management terminal 15 as a notification of the conduction confirmation result of the new path. In the result display window WD4, the result of continuity of the path with the iSNS server 50 through each IP port 110 is displayed. When it is determined that the conduction result is not within the allowable range (step S740), it may be displayed that the iSNS server address is not registered.

  Further, the program for diagnosing the IP network based on the ICMP protocol stored in advance in the ROM 212 of the storage control unit 20 is not limited to ping, but may be traceroute or tracert. In this embodiment, the storage unit 30 is configured by a plurality of HDDs 310. However, the storage unit 30 may be configured by various types of storage such as an optical disk drive, a semiconductor memory, and a magnetic tape drive.

  Further, in the path recovery process shown in FIG. 15, the path is recovered by transferring the port address of the failed port or the high load port to an unused port. The path may be restored by assigning a port address. FIG. 20 is a flowchart illustrating a path restoration process according to another embodiment. FIG. 21 is an explanatory diagram illustrating an example of a port address table according to another embodiment. In the path recovery process shown in FIG. 20, the target computer 20 refers to the IP port communication statistics table 16000 shown in FIG. 16 instead of searching for unused ports (step 430 in FIG. 15), thereby detecting the failure / high A low load port whose load factor does not exceed a predetermined value is searched from IP ports 110 other than the load port (step S630). In this embodiment, if the load factor does not exceed 40%, it is determined as a low load port. If the target computer 20 has a low load port (step S635), the target computer 20 replaces the fault path / high load port of the existing path with the low load port based on the existing path using the fault / high load port. Then, it is set as a conduction path for conducting conduction confirmation (step S640). If continuity is normal in the continuity confirmation process (step 450 in FIG. 15) executed thereafter, the target computer 20 adds the port address assigned to the fault / high load port to the low load port. Then, the port address table 21000 shown in FIG. 21 is rewritten (step S670). A port address table 21000 illustrated in FIG. 21 includes information on IP port names 21010 of the IP ports 110 and port addresses 21020, 21030, and 21040 assigned to the IP ports 110. In the example of the port address table 21000 shown in FIG. 21, it is possible to assign up to three port addresses to one IP port 110, but the number of port addresses assigned to one IP port 110 Is not limited to three, and can be appropriately increased or decreased according to the embodiment. When the load on the IP port 110 to which a plurality of port addresses are assigned increases, some of the plurality of port addresses may be deleted. According to the path restoration process shown in FIG. 20, the load can be distributed among the plurality of IP ports 110.

  Further, the connection of each device such as the storage system 10, the initiator computer 70, and the iSNS server 50 is not limited to the connection mode shown in FIG. 1 of this embodiment, and various connection modes can be adopted. For example, the storage system 10, the initiator computer 70, and the iSNS server 50 are connected by an IP-SAN 60 and an out-of-band network independent from the IP-SAN 60 so that traffic of the IP-SAN 60 is not mixed. It is also good to do. FIG. 24 is an explanatory diagram showing a physical configuration of an electronic computer system CS in another embodiment. In the example of the connection mode shown in FIG. 24, the storage system 10, the initiator computer 70, and the iSNS server 50 are connected via an out-of-band network 80 different from the IP-SAN 60. Various settings in the initiator computer 70 are performed by the iSNS server 50 via the management network 80. In the connection mode illustrated in FIG. 24, the management terminal 15 may be connected to the storage system 10 and various settings of the storage system 10 may be performed by the iSNS server 50 and the management terminal 15. Further, regarding the paths between the devices in the management network 80, the continuity confirmation may be performed in the same manner as the continuity confirmation processing shown in FIG. 7 and FIG.

It is explanatory drawing which shows the physical structure of the electronic computer system CS. 2 is a block diagram mainly showing an internal configuration of a storage system 10. FIG. It is explanatory drawing which shows the logical structure of the electronic computer system CS. It is explanatory drawing which shows an example of a target setting table. It is explanatory drawing which shows an example of a path | pass setting table. It is a flowchart which shows a new path setting process. It is a flowchart which shows a continuity confirmation process. It is explanatory drawing which shows an example of the mode of a continuity confirmation process. It is explanatory drawing which shows an example of the notification of the conduction confirmation result of a new path | pass. It is a flowchart which shows an initiator information acquisition process. It is explanatory drawing which shows an example of a storage domain setting table. It is explanatory drawing which shows an example of an initiator address table. It is a flowchart which shows an existing path confirmation process. It is explanatory drawing which shows an example of the notification of the conduction confirmation result of the existing path. It is a flowchart which shows a path | pass recovery process. It is explanatory drawing which shows an example of an IP port communication statistics table. It is explanatory drawing which shows an example of a port address table. It is explanatory drawing which shows an example of the notification of a path | pass recovery process result. It is a flowchart which shows the conduction confirmation process of other embodiment. It is a flowchart which shows the path | pass recovery process of other embodiment. It is explanatory drawing which shows an example of the port address table of other embodiment. It is a flowchart which shows a server connection setting process. It is explanatory drawing which shows an example of the notification of a server connection setting result. It is explanatory drawing which shows the physical structure of the electronic computer system CS in other embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Storage system 15 ... Management terminal 16 ... Display 17 ... Keyboard 20 ... Storage control part 20 ... Target computer 30 ... Storage part 50 ... iSNS server 55 .. Management terminal 56 ... Display 57 ... Keyboard 60 ... IP-SAN
70 ... Initiator computer 80 ... Management network 110 ... IP port 130 ... LAN port 210 ... CPU
212 ... ROM
214 ... RAM
220 ... Data controller 230 ... Cache memory 310 ... HDD
CS ... Electronic computer system

Claims (28)

A storage system comprising: a plurality of connection ports connectable to a network based on an Internet protocol; and a target computer that provides a predetermined storage volume through at least one of the plurality of connection ports to an initiator computer connected to the network In the conduction confirmation method for confirming conduction between the initiator computer and the target computer,
An echo requesting step of transmitting an echo request based on an Internet control message protocol by the target computer to the initiator computer through a predetermined connection port of the plurality of connection ports;
Based on the reception state at the target computer of the echo response returned from the initiator computer as a response to the transmitted echo request, between the initiator computer and the target computer through the predetermined connection port by the target computer A continuity determination step for determining whether or not the continuity is normal. A conduction confirmation method according to claim 1,
The predetermined connection port is a connection port that is newly set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports,
Further, when it is determined that the continuity between the initiator computer and the target computer through the predetermined connection port is normal, the setting for using the connection port as the data transfer path is registered by the target computer. A conduction confirmation method including a path setting registration process. A conduction confirmation method according to claim 1,
The predetermined connection port is a connection port different from a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports,
Further, when a communication failure has occurred in the connection port set as the data transfer path, the connection port set as the data transfer path is determined to be normal by the target computer as the data transfer path. A continuity check method including a path setting change step for changing the setting to at least one of the ports. A conduction confirmation method according to claim 1,
The predetermined connection port is a connection port different from a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports,
Further, when the communication load of the connection port set as the data transfer path exceeds a predetermined value, the connection port set as the data transfer path is normally connected to the connection port set by the target computer. A continuity check method comprising a path setting change step for changing the setting to at least one of the determined connection ports.   Further, when changing the setting of the connection port in the path setting changing step, a port address indicating the location of the connection port on the network, which has been assigned to the connection port of the setting change source, is used as the connection port of the setting change destination. The continuity confirmation method according to claim 3 or 4, further comprising an address transfer step of transferring. A conduction confirmation method according to claim 1,
The predetermined connection port is a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports,
The target computer is capable of exchanging data with a management terminal that exchanges predetermined information with an administrator of the storage system,
Further, based on an input of an instruction from the administrator via the management terminal by the target computer, a determination result as to whether or not the continuity for the predetermined connection port is normal is displayed on the management terminal. A continuity confirmation method comprising a determination result notifying step of notifying the manager via the determination result.   The continuity determination method according to claim 1, wherein the continuity determining step is a step of determining that the continuity is normal when the echo response reaches the target computer.   The continuity determining step is a step of determining that the continuity is normal when an arrival rate at which the echo response reaches the target computer exceeds a predetermined value with respect to a plurality of echo requests. 6. The conduction confirmation method according to any one of 6. The continuity confirmation method according to any one of claims 1 to 8,
The target computer can exchange data with a user output interface that outputs predetermined information to an administrator of the storage system,
Further, when it is determined that the continuity between the initiator computer and the target computer through the predetermined connection port is abnormal, the reason for the abnormality is determined by the target computer via the user output interface. Continuity confirmation method provided with abnormality notification process of notifying to A conduction confirmation method according to any one of claims 1 to 9,
The target computer is capable of exchanging data with a user input interface that receives input of predetermined information from an administrator of the storage system,
The continuity confirmation method further comprising an address acquisition step of acquiring an initiator address indicating the location of the initiator on the network from the administrator via the user input interface by the target computer. A conduction confirmation method according to any one of claims 1 to 9,
The target computer receives an input of predetermined information from a storage name server having an initiator address indicating the location of the initiator on the network in association with an initiator name that can identify the initiator computer, and an administrator of the storage system Data can be exchanged with the user input interface.
Further, the target computer acquires the initiator name from the administrator via the user input interface, and acquires the initiator address held by the storage name server in association with the acquired initiator name from the storage name server. A continuity confirmation method including an address acquisition step. A conduction confirmation method according to any one of claims 1 to 9,
The target computer includes a storage domain defining a combination of the initiator computer and the target computer, and a storage name server holding an initiator address indicating the location of the initiator computer defined in the storage domain on the network; Can exchange data,
The continuity check method further comprising an address acquisition step of acquiring, from the storage name server, the initiator address of the initiator computer defined in the storage domain as a combination with the target computer by the target computer.   The continuity confirmation method according to claim 1, wherein at least one of the initiator computer and the target computer is plural.   The continuity confirmation method according to any one of claims 1 to 13, wherein communication between the initiator computer and the target computer is communication based on an iSCSI protocol. A storage system comprising: a plurality of connection ports connectable to a network based on an Internet protocol; and a target computer that provides a predetermined storage volume through at least one of the plurality of connection ports to an initiator computer connected to the network Because
The target computer is
Echo request means for transmitting an echo request based on the Internet control message protocol to the initiator computer through a predetermined connection port of the plurality of connection ports;
Based on the reception state of the echo response sent back from the initiator computer as a response to the transmitted echo request, the continuity between the initiator computer and the target computer through the predetermined connection port is normal. A storage system comprising continuity determination means for determining whether or not there is. The storage system according to claim 15, wherein
The predetermined connection port is a connection port that is newly set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports,
The target computer further registers a setting for using the connection port as the data transfer path when it is determined that conduction between the initiator computer and the target computer through the predetermined connection port is normal. A storage system provided with path setting registration means. The storage system according to claim 15, wherein
The predetermined connection port is a connection port different from a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports,
The target computer further determines that the connection port set as the data transfer path is normal when the communication failure occurs in the connection port set as the data transfer path. A storage system comprising path setting change means for changing the setting to at least one of the ports. The storage system according to claim 15, wherein
The predetermined connection port is a connection port different from a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports,
When the communication load of the connection port set as the data transfer path exceeds a predetermined value, the target computer further connects the connection port set as the data transfer path with the normal conduction. A storage system comprising path setting change means for changing the setting to at least one of the determined connection ports. The target computer further changes the port address indicating the location of the connection port on the network that was assigned to the connection port of the setting change source when the connection port setting is changed by the path setting changing unit. The storage system according to claim 17 or 18, further comprising address transfer means for transferring to a previous connection port. The storage system according to claim 15, wherein
The predetermined connection port is a connection port set as a data transfer path between the initiator computer and the target computer among the plurality of connection ports,
The target computer is
It is possible to exchange data with a management terminal that exchanges predetermined information with an administrator of the storage system,
Further, based on an instruction input from the administrator via the management terminal, a determination result on whether or not the continuity for the predetermined connection port is normal is sent to the administrator via the management terminal. A storage system provided with a determination result notifying means for notifying to a storage system.   21. The storage system according to claim 15, wherein the continuity determination unit is a unit that determines that the continuity is normal when the echo response reaches the target computer.   The continuity determining means is a step of determining that the continuity is normal when an arrival rate at which the echo response reaches the target computer exceeds a predetermined value with respect to a plurality of echo requests. 20. The storage system according to any one of 20. A storage system according to any one of claims 15 to 22,
The target computer is
It is possible to exchange data with a user output interface that outputs predetermined information to the administrator of the storage system,
Further, when it is determined that the continuity between the initiator computer and the target computer through the predetermined connection port is abnormal, the reason for the abnormality is notified to the administrator via the user output interface. A storage system with an error notification means. The storage system according to any one of claims 15 to 23, wherein
The target computer is
It is possible to exchange data with a user input interface that receives input of predetermined information from the administrator of the storage system,
Furthermore, a storage system comprising address acquisition means for acquiring an initiator address indicating the location of the initiator on the network from the administrator via the user input interface. The storage system according to any one of claims 15 to 23, wherein
The target computer is
A storage name server that stores an initiator address indicating the location of the initiator on the network in association with an initiator name that can identify the initiator computer, and a user input interface that receives input of predetermined information from an administrator of the storage system While exchanging data,
Further, an address acquisition means for acquiring the initiator name from the administrator via the user input interface and acquiring the initiator address held by the storage name server in association with the acquired initiator name from the storage name server. Storage system provided. The storage system according to any one of claims 15 to 23, wherein
The target computer is
Data exchange with a storage domain defining a combination of the initiator computer and the target computer, and a storage name server having an initiator address indicating the location of the initiator computer on the network defined in the storage domain Is possible,
Furthermore, a storage system comprising address acquisition means for acquiring the initiator address of the initiator computer defined in the storage domain as a combination with the target computer from the storage name server.   27. The storage system according to claim 15, wherein at least one of the initiator computer and the target computer is plural.   28. The storage system according to claim 15, wherein communication between the initiator computer and the target computer is communication based on an iSCSI protocol.

Images